Pankaj Javale

Bio: Pankaj Javale is an academic researcher. The author has contributed to research in topics: Wireless network & Image compression. The author has an hindex of 1, co-authored 1 publications receiving 3 citations.

Performance evaluation of wireless image transmission: An approach using embedded system

Abstract: For the latest development in the direction of embedded system and wireless network, this project work describes wireless image transmission based on the embedded system. The Project Work Includes Following Features of Image Transmission using WIFI from S3C2440A to PC, S3C2440A is the h/w, with QTOPIA OS, PC with LINUX(Ubuntu) and with Socket code being in C/C++ language. Here S3C2440A will be working as Server and PC with Linux Platform(UBUNTU) will be working as Client. Experimental results show the throughput, transmission time and delay evaluation for various image file types along with speed gain using image compression.

Networked hardware assisted key image and chaotic attractors for secure RGB image communication

Abstract: In multimedia communication, significance of the images for data representation is noteworthy. In this context, secure transmission of images over open channel has become a challenging task. Creation of different strategies in improving the secure image transmission always has a demand. The proposed work suggests an RGB image encryption with the confluence of attractors and hardware triggered key image in which confusion and diffusion were accomplished by Lorenz, Lu and Cellular Automata attractors. The uniqueness of proposed encryption scheme is a key image generation module through cascaded Ring Oscillator circuit which creates M × N key image for diffusion of pixels. Facilitating the authenticated networked access to key image generation hardware enables the secure server-client architecture for a variety of secure image transfer applications. The proposed approach is a hardware – software codesign which possesses a good keyspace, improved key sensitivity and satisfies the various statistical parameters thus offering substantial resistance to differential, occlusion and chosen plaintext attacks on RGB images.

Transmission of Multimedia Data in Underwater Terrain using Acoustic Waves

Abstract: Underwater acoustic networking is a new technology tool that can be used for oceanographic collection of data, emissions control, off shore exploration, and navigation. Communication in UWSNs is challenging and transmission of multimedia data becomes more difficult due to the peculiar features of the underwater network. During military emergency transferring multimedia data becomes crucial so as to get the proper knowledge about the situation. Therefore, this paper has proposed a simulation based mechanism for transmitting multimedia data in an underwater terrain. Acoustic waves have been used for the transmission. Multimedia data includes images and videos and for that we have taken four scenarios comprises of low quality image, high quality image, low quality video and high quality video. The simulation result shows that low quality image is most suitable in the underwater terrain.

Deep-Learning-Based Wireless Visual Sensor System for Shiitake Mushroom Sorting

Abstract: The shiitake mushroom is the second-largest edible mushroom in the world, with a high nutritional and medicinal value. The surface texture of shiitake mushrooms can be quite different due to different growing environments, consequently leading to fluctuating market prices. To maximize the economic profit of the mushroom industry, it is necessary to sort the harvested mushrooms according to their qualities. This paper aimed to develop a deep-learning-based wireless visual sensor system for shiitake mushroom sorting, in which the visual detection was realized by the collection of images and cooperative transmission with the help of visual sensors and Wi-Fi modules, respectively. The model training process was achieved using Vision Transformer, then three data-augmentation methods, which were Random Erasing, RandAugment, and Label Smoothing, were applied under the premise of a small sample dataset. The training result of the final model turned out nearly perfect, with an accuracy rate reaching 99.2%. Meanwhile, the actual mushroom-sorting work using the developed system obtained an accuracy of 98.53%, with an 8.7 ms processing time for every single image. The results showed that the system could efficiently complete the sorting of shiitake mushrooms with a stable and high accuracy. In addition, the system could be extended for other sorting tasks based on visual features. It is also possible to combine binocular vision and multisensor technology with the current system to deal with sorting work that requires a higher accuracy and minor feature identification.